Flood frequency analysis is a technique utilized to evaluate the probability and magnitude of flood occurrences at specific river locations. In the present study, 50 gauging stations were carefully selected across the Godavari river basin to perform flood frequency analyses. In the present study, flood frequency analysis incorporates two distinct approaches: local flood frequency analysis (LFFA) and regional flood frequency analysis (RFFA). These approaches are applied to analyse two types of data series: annual maximum (AM) series and peaks over threshold (POT) series. LFFA focuses on a specific location or site along a river. On the other hand, regional flood frequency analysis takes a broader perspective and considers multiple locations or sites within a homogeneous region. In the case of the AM series, LFFA and RFFA parameters were estimated using the Bayesian method. For the POT series, the maximum likelihood approach was employed to estimate shape, scale, and local parameters. In the RFFA, parameters estimated from multiple stations within the region were pooled together using the Index flood method. These pooled parameters were then used to determine flood quantiles at ungauged locations within the region. The analysis of the AM series revealed a combination of significant positive and negative trends at various stations, reflecting intricate flow pattern variations. On the other hand, the POT approach displayed fewer notable trends, primarily leaning towards negative changes. Interestingly, stations located in the middle of the basin demonstrated increasing flow rates over time, suggesting potential shifts in hydrological dynamics. The results showed RFFA consistently yielded higher quantile estimates compared to LFFA, signifying the importance of considering regional perspectives. The findings of this study provide valuable insights for effective flood management and mitigation strategies in Godavari Basin.